No, the file needs to be nothing but hex value 55 throughout. The
attachment I sent earlier is just that:
smarlowe(at)sawyer:~$ hexdump 55
0000000 5555 5555 5555 5555 5555 5555 5555 5555
i.e. all 55s, all the time, not the ascii numbers 5 and 5, but the hex
value. The actual file is ascii the letter U over and over:
smarlowe(at)sawyer:~$ head -c 100 55
and so on.
On Wed, Sep 24, 2008 at 12:09 PM, Carol Walter <walterc(at)indiana(dot)edu> wrote:
> To use the hex value 0x55, do I need to enclose it in single quotes?
> On Sep 24, 2008, at 11:44 AM, Tom Lane wrote:
>> Carol Walter <walterc(at)indiana(dot)edu> writes:
>>> Are the files that contain the hex characters supposed to contain a
>>> single string and no control characters?
>> Yes, you want 256K occurrences of the byte value 0x55 and nothing else.
>>> I'm also wondering if, after I create the dummy files, and pg_dump
>>> works, I could restore an old pg_dumpall file and then insert any
>>> data that aren't there from the pg_dumps.
>> It'd be a good idea to do as much cross-checking as you can, since
>> it's highly probable that the dumped data will be at least partly wrong.
>> No, I'm afraid updating to 8.2.latest won't get you out of this.
>> It might possibly prevent a recurrence.
>> regards, tom lane
>> Sent via pgsql-admin mailing list (pgsql-admin(at)postgresql(dot)org)
>> To make changes to your subscription:
In response to
pgsql-admin by date
|Next:||From: Tom Lane||Date: 2008-09-24 20:26:39|
|Subject: Re: missing chunk number 0 for toast value |
|Previous:||From: Scott Marlowe||Date: 2008-09-24 19:43:51|
|Subject: Re: Hex representation|